AG

AA

bst

ract

srecruitment and that AT-1001 could delay progression of enteropathy in patients withgluten sensitivity.

T1270

Gliadin Induced Increase in Intestinal Epithelial Permeability Is Independentof MEKAlex Staudacher, Barry C. Powell, Guy R. Sander

Background: A characteristic of coeliac disease during ingestion of gluten is increased intest-inal permeability, dysregulation of junctional proteins located at the apical lateral membraneand phosphorylation of the extracellular signal-regulated kinase proteins ERK1 and ERK2(ERK1/2). The aim of this study was to test whether gliadin induced phosphorylation ofERK1/2 proteins contributes to increased permeability. Methods: We challenged establishedconfluent T84 monolayers grown on Transwells with pepsin and trypsin digested gliadin(1mg/ml; PT-gliadin) for 4 hours. Parallel T84 cells were pre-incubated with the MEKinhibitor (PD98509; 50 µM) for 1 hour before gliadin challenge. Permeability was analysedby measuring transepithelial resistance (TER) and influx of 4 kDa FITC-dextran. Tightjunction protein expression and ERK1/2 phosphorylation were detected by Western blotanalysis and intracellular relocation of tight junction proteins was detected by confocalmicroscopy. Results: Gliadin increased ionic permeability of T84 monolayers rapidly. A 96%decrease in TER occurred within 30 minutes of PT-gliadin challenge (172 ± 6 Ohms/cm2)when compared to T84 monolayers treated with BSA (4,535 ± 33 Ohms/cm2). A statisticallysignificant increase in the influx of 4 kDa FITC dextran was first detected at 60 minutes(p<0.015) in monolayers treated with PT-gliadin (1.824 ± 0.052 µg/cm2) compared tomonolayers treated with BSA (0.073 ± 0.014 µg/cm2). Increased permeability coincidedwith intracellular re-distribution of claudin -1, -3, -4 and -5 proteins from the lateralmembrane to the cytoplasm but not occludin or ZO-1. Western blot analysis revealed thatthe overall level of expression of claudin-1, -3, -4, -5 and occludin did not change. PT-gliadin induced an increase in the phosphorylation of ERK1/2 proteins (44 kDa and 42kDa) within 10 minutes correlating with an increase in permeability. The MEK inhibitorprevented PT-gliadin-induced ERK1/2 phosphorylation but did not prevent an increase inpermeability. Conclusion: Our findings show that PT-gliadin induced increase in permeabilityin T84 intestinal cells is MEK independent.

T1271

Enhanced Levels of Interleukin -1beta in Stools from Children withInflammatory Bowel Disease, But No Local Production in Children withCoeliac DiseaseTony Hansson, Helene Lettesjo, Christer G. Peterson, Britt-Inger Nyberg, Ingrid Dahlbom,Anders Dannaeus

Background: Interleukin-1beta (IL-1beta) is a proinflammatory cytokine that induces feverand acute phase protein synthesis. In active inflammatory bowel disease (IBD) large amountsof IL-1beta is produced by mucosal macrophages and increased amounts of IL-1 beta hasbeen found in the rectal mucosal fluid. IL-18 is a pro-inflammatory cytokine that has acrucial role in maintaining the T helper cell type 1 (Th1) response. IL-18 is, together withother Th1 cytokines, expressed in the small intestinal mucosa of patients with coeliac disease(CoD). Aim: The aim of this study was to compare the levels of IL-1beta and IL-18 inchildren with CoD, IBD and other gastrointestinal diseases. Methods: IL-1beta were measuredin stool samples and IL-18 in serum samples from 172 children with gastrointestinal com-plaints (median age 12 years). Included were 17 untreated children with CoD; 35 CoDchildren on a gluten-free diet; 40 children with IBD; 40 children with food hypersensitivity,and 40 children with other diseases. Results: Enhanced levels of IL-1beta were found instool samples from 20 children and 16 of them had an active IBD (median 13 ng/g). Onlyone celiac child had slightly increased IL-1beta levels (1.8 ng/g) and the median level forthe 17 untreated celiac children were 0.37 ng/g. The untreated CoD children had the highestmedian level of IL-18 (259 pg/ml). However, of the 17 children with increased IL-18 levels,4 had CoD, 6 food hypersensitivity and 6 had other gastrointestinal complaints. Notably,only one IBD patient had increased levels of IL-18, a child with Crohns disease, who inaddition was the only child with elevated concentrations of both IL-1beta (152 ng/g) andIL-18 (663 pg/ml). Conclusion: The production of intestinal IL-1beta and peripheral IL-18did not coincide and the results indicate that separate inflammatory mechanisms are involvedin the disease development of CoD and IBD. The different cytokine expression can alsoelucidate why IBD patients often suffer from fever, while patients with CoD or food hypersens-itivity seldom are affected.

T1272

Role of NKT Cell-Driven IFN-γ in Postoperative Adhesion FormationHisashi Kosaka, Tomohiro Yoshimoto, Kenji Nakanishi, Jiro Fujimoto

Background: Adhesion formation is a common and often severe complication of abdominalsurgery. However, the mechanism of adhesion formation is still poorly understood. The aimof this study is to establish new experimental mouse model of surgical adhesion formationand to examine immunological mechanisms underlying organ adhesions. Method: Anesthet-ized mice were underwent an operation for cauterization of cecum by coagulate mode ofsurgical forceps for a second. Each mouse was sacrificed 7days later and evaluated degreeof adhesion formation following standard scoring system (Score from 1 to 5), which hasbeen widely used in this field. Cecums were prepared for measurement of mRNA expressionof cytokines by real time PCR. Result: 1.All animals survived and developed severe adhesionformation (Score 5; very thick adhesion). Histological analysis of cecum exhibited severeinflammation and fibrosis. 2.To evaluate the role of CD4+T cells for surgical adhesionformation, mice were depleted of CD4+T cells by treatment with anti-CD4 antibody. TheseCD4+T-depleted mice formed mild or no adhesion formation (Score 0-1). 3.The jα18 KOmice lacking NKT cells also formed mild or no adhesion formation (Score 0-1). 4.Todetermine which cytokines mediate this surgical adhesion formation, we isolated cecums

T : 11501$$CH204-02-08 16:47:11 Page 520Layout: 11501B : e

A-520AGA Abstracts

and examined mRNA expression. In general, cytokine mRNAs were undetectable in cecum,however, IFN-γ mRNA expression increased and peaked at 3 h after operation. 5.IFN-γ KOmice formed mild or no adhesion formation (Score 0-1). However, IFN-γ KO mice givenIFN-γ formed severe adhesion formation (Score 5). Furthermore, Wild-type mice given antiIFN-γ antibody formed mild adhesion formation (Score 2). 6.To evaluate the significanceof IFN-γ from NKT cells, we transferred NKT cells from wild-type or IFN-γ KO mice tojα18 KO mouse. Only jα18 KO mouse given wild-type NKT cells formed severe adhesionformation (Score 5). 7.Since development of Th1 cells requires IL-12, we compared thedegree of adhesion formation between STAT4 KO and STAT6 KO mice. Both mice formedsevere adhesion formation (Score 5), excluding contribution of Th1 or Th2 cells to develop-ment of adhesion. In contrast, STAT1 KO mouse formed mild or no adhesion formation,substantiating further that IFN-γ is a causative factor for this surgical adhesion formation.Conclusion: These studies demonstrated the immunological mechanism underlying surgicaladhesion formation. Contribution of IFN-γ from NKT cells to adhesion development shedsa new insight in understanding the mechanism of adhesion formation and provides targetcells or molecules for regulation of organ adhesion.

T1273

CD30 Antigen Expression Is Involved in Celiac DiseaseNatalia Periolo, Laura Guillen, Marcos Barboza, Sonia Niveloni, Eduardo Mauriño, JulioC. Bai, Alejandra C. Cherñavsky

Background/aims: CD30 antigen is expressed after activation of normal B and T lymphoidcells. Different T cell activating stimuli induce CD30 expression on CD45RO+ precursors.The putative role of CD30 in celiac disease (Cel) is poorly known. Our aims are to determineCD30 expression after nonspecific activation of peripheral blood lymphocytes (PBL) and toexamine CD30 expression in duodenal T cells from biopsies ex-vivo challenged with gliadin.Materials/methods: PBL from 10 healthy controls (Co) and 10 active adult Cel patients wereincubated with 1µg/ml of anti-CD3 for 5 days. Triple immunofluorescence (anti-CD45ROPE or -CD25 PE, -CD3 PerCP and -CD30 FITC) and flow cytometry analysis were performedat baseline and days 3 and 5 after incubation. CD30 antigen was immunohistochemicallyevaluated in paraffin sections of duodenal biopsies from 6 Cel patients and 6 Co. Tripleimmunofluorescence and flow cytometry analysis was performed on isolated intraepithelialand lamina propria lymphocytes (IEL and LPL). Biopsy specimens from 3 Co and 3 Celwere similarly evaluated after 3-hour incubation of biopsies with 100 µg/ml of crude gliadin.Results: Meanwhile CD30 is not expressed on resting PBL, a peak is shown after 3-dayincubation with anti-CD3 (Cel vs Co, p=ns). Compared with Co samples, CD30 expressionpersists increased by day 5 in Cel (1.2±1.0 vs.13.3±3.0, p<0.05). CD30 frequency is increasedon CD3+CD45RO+ and CD3+CD25+ subsets (10.5±2.0% vs. 0.3±0.1%, p=0.0238 and11.1±1.9% vs. 0.6±0.2%, p=0.0025, Cel vs Co, respectively). While a similar low frequencyof CD30+ cells was found at baseline in isolated IEL from Cel and Co (p=ns), the expressionwas increased in Cel LPL (vs. Co: 12.4±5.1% and 7.9±3.3%, p=0.0367). Incubation withgliadin up-regulates CD30 expression in a subpopulation of LPL from both, Co and Cel.Conclusions: Cel patients show a persistent expression of CD30 on a subset of memory,activated cells that might be capable of signal transduction and differential immunoregulatoryactivity in peripheral compartment. Similarly, the higher frequency of CD3+CD30+ LPLsobserved points to the presence of a differential activated subset of duodenal T cells probablyinvolved in CD pathogenesis within the intestinal mucosa. After challenging of biopsies withgliadin, CD30 triggering might provide costimulatory signals for activation/proliferation ofLPLs in both patients and controls. However, a functionally differential response remainsto be demonstrated.

T1274

Dominant Mucosa-Associated Microbiota in Celiac Children At Diagnosis andAfter GFDValerio Iebba, Maria Pia Conte, Maria Barbato, Serena Schippa, Osvaldo Borrelli, CatiaLonghi, Giulia Maiella, Valentina Totino, Franca Viola, Salvatore Cucchiara

Background Celiac disease (CD) is an immune-mediated enteropathy, characterized by smallbowel chronic inflammation. Its pathogenesis is multifactorial: exposure to toxic prolaminsand appropriate HLA-DQ haplotype are necessary but not sufficient for contracting CD.Althoug modification of gut microbiota seems to be involved in the pathogenesis of otherchronic inflammatory bowel diseases (IBD), its possible role in CD has never been investig-ated. Aims To identify by a molecular approach the microbiota colonizing the upper smallbowel of children with CD at diagnosis and after 8 mounths of gluten free diet (GFD).Methods Mucosal-associated bacteria from duodenal biopsies of 10 celiac children aged 5-15 years ,(at diagnosis and after 8 mounths of GFD), and of 8 healthy controls wereinvestigated. Total DNA was extracted, and 16S ribosomal DNA was amplified by PCR.Amplification products were separated by temporal temperature gradient gel electrophoresis(TTGE) a powerful technique for comparing the biodiversity of the dominant microbiotain different biological samples. The profiles obtained were analyzed using GelQuest soft-ware(Sequentix) providing a dendogram based on the agglomeration method of UPGMA.Result The dendogram obtained reveals two clusters sharing high degrees of similarity. Thefirst cluster includes all active disease patients, the second one those in GFD. TTGE profilesof controls clustered with celiac children in GFD. Interestingly dominant microbiota inactive disease differed very much from remission and controls. Conclusion This is the firstpediatric report investigating the duodenal mucosa-associated microbiota in celiac childrenat diagnosis and after GFD. The main result of the present study, although limited by thesample size, hjghlighted TTGE profiles clusters with clinical status of CD. These preliminarydata showed that dominant duodenal microbiota in active disease seems to differ very muchfrom that in remission. These findings suggest a pathogenetic role in CD. Further studieswill sequence different DNA fragments, specific for each profile, to identify dominant micro-bial species characterizing intestinal microbiota in active and in remission disease.